Anastomotic failure due to acute vasculitis in free flap transfer: A case report

We report a case of anastomotic failure secondary to acute vasculitis in a free flap. Following transfer of a latissimus dorsi myocutaneous flap, massive hemorrhage from the anastomosis occurred six days postoperatively. This was associated with segmental infarctions of the areas of the flap, which was removed. Transmural necrosis of the thoracodorsal artery, secondary to acute vasculitis was demonstrated by histology. Subsequent biopsies revealed an autoimmune vasculitis. MICROSURGERY 13:7-10

1992

ANASTOMOTIC FAILURE DUE TO ACUTE VASCULITIS IN FREE FLAP TRANSFER: A CASE REPORT JAMES R. SANGER, M.D., BRUCE 8. KADZ, M.D., THU MlNH TIEU, M.D., HANl S. MATLOUB, M.D., and N. JOHN YOUSIF, Y.D.

M o s t free flap failures are caused by thrombosis of either the arterial or venous anastomosis.’ The majority of these are secondary to technical problems at the time of anastomosis or compression of the vascular pedicle in the postoperative period. Unusual causes, including coagulation defects, occasionally occur. We report a case in which acute vasculitis resulted in disruption of the anastomosis and subsequent free flap loss. CASE REPORT

A 57-year-old white male was seen for treatment of bilateral lower leg ulcers. The patient had had extensive skin and subcutaneous tissue loss over 70% of each leg. Several months prior he became acutely ill. He developed severe hypotension, peripheral edema, and ecchymotic skin lesions on the legs. He was treated for sepsis with broad spectrum antibiotics and subsequent cultures revealed Staphylococcus species and Klebsiella species from sputum and the skin lesions of the leg. The ecchymotic lesions progressed to full-thickness skin necrosis. An extensive medical work-up revealed evidence of autoimmune disease. A kidney biopsy showed a membranoproliferative glomerulonephritis. Skin biopsies revealed a vasculitis. At an outside facility, skin grafting was done following debridement to healthy underlying muscle and fascia. The skin grafts failed. There was exposure of the left achilles tendon and loss of the peritenon. Extensive ulceration over a portion of the ankle was also present. From the Department of Plastic and Reconstructive Surgery (J.R.S., H.S.M., N.J.Y.); Department of Surgery (B.B.K.); and the Department of Pathology (T.M.T.), Medical College of Wisconsin, Milwaukee, WI. Address reprint requests to James R. Sanger, M.D., 9200 W. Wisconsin Avenue, Milwaukee, WI 53226. Received November 29, 1990; revision accepted June 26,1991.

0 1992 Wiley-Liss, Inc.

The patient was transferred to our facility and evaluated by the medical service. He was felt to be stable medically and was prepared for surgery in attempt to salvage the leg. Corticosteroid therapy was not recommended. A latissimus dorsi myocutaneous flap was transferred to the lower extremity, using end-to-side anastomosis of the thoracordorsal artery to the posterior tibia1 artery and an end-to-end anastomosis of the thoracodorsal vein to the venae comitante. On the second postoperative day, bleeding was noted at the edge of the flap. At the bedside, a small hematoma was evacuated and the bleeding stopped. The skin paddle and muscle appeared healthy. Recurrent bleeding on the fourth postoperative day required return to the operating room, where disruption of a portion of the arterial anastomosis was noted. Revision of the arterial anastomosis was carried out. Subsequent to this, patchy areas of demarcation developed within the flap. Although some areas of the flap bled normally, some areas did not bleed at all. Massive hemorrhage from the arterial anastomosis occurred on the sixth postoperative day and he was returned to surgery, where the flap was removed. The patient’s hospital course was one of steady downhill progression. He became disorientated and developed lung infiltrates and pulmonary edema. Repeat skin biopsies revealed acute vasculitis with heavy deposits of compliment and IgM. A small nasal septal defect was noted. It was felt that the combination of the nasal septal defect, golmerulonephritis, pulmonary infiltrates and vasculitis was consistent with Wegener’s granulomatosis.* Cytoxan and methylprednisone were instituted. He initially appeared to improve, but 48 hours after institution of cytoxan and methylprednisone, and 3 weeks following his attempted free flap, he developed circulatory collapse and died. Autopsy revealed widespread acute vasculitis and multiple mycotic abscesses involving the heart and lung. Other septic emboli

8

Sanger et al.

Figure 7. The arterial tissue recovered from the anastomosis demonstrated transmural inflammatory infiltrates composed mainly of polymorphonuclear neutrophilic leukocytes. In addition, focal calcification is also identified. Stain for fungus was negative. “L” designates the lumen. (Original magnification, 250 x )

were noted. Paecilomyces species, an opportunistic organism, was recovered from several abscess sites. Histology of the anastomosis from the resected free flap revealed acute vasculitis (Figs. 1, 2). Areas of segmental infarction were apparent (Fig. 3). No evidence of a mycotic aneurysm or fungal organisms was detected at the site of anastomosis. DISCUSSION

Failures in free flap transfer relate primarily to vascular occlusive problems. By far the majority of problems relate to endothelial damage and technical problems, either with the anastomosis or in the placement of the pedicle, which predisposes it to compression. In most series, arterial problems have been predominant, although venous complications have bzen equally prevalent in our series of over 400 flaps. Hematoma may compromise flow and cause thrombosis. Infections adjacent to vascular repairs can cause failure due to thrombosis or erosion of the vessels. Previous

Figure 2. A higher magnification of the vessel wall identifies the acute inflammatory process. (Original magnification, 500 x )

trauma, irradiation, and atheromatous damage of vessels are also contributing factors. Wegener’s granulomatosis is a disease of unknown etiology, although believed to be of autoimmune origin.* It predominantly affects men in the 40 to 50 age-group, and the presenting symptoms may include sinusitis, nasal mucosal ulcerations, purulent otitis media, henioptysis, and pleuritis. Renal involvement manifested by proteinuria, hematuria, and erythrocyte casts is common. Less commonly, skin vasculitis has been noted. Certain risk factors, which predispose to thrombotic abnormalities have been associated with autoimmune vasculitis. Vasculitis is simply defined as an inflammation directed against vessels. Although little is known about etiologic agents that contribute to pathogenesis of vasculitis, many types of vasculitis are thought to be induced by immunologic mechanisms. Initial studies linking vasculitis to an imnrunologic process were performed in rabbits by Rich and G r e g ~ r yThis .~ concept was further refined by Dixon et al. ,4 who suggested that the vascular destruction could be caused by circulating small immune complexes. This was expanded upon by

Atypical Anastornotic Failure

9

Figure 3. Free flap at time of removal. Note areas of segmental infarction.

Smoller et al.,5 who explained that immune complexes in the vessel walls serve to activate the complement cascade that chemotactically attracts polymorphonuclear leukocytes. During the attempt to phagocytize these immune complexes, lysosomal enzymes are released extracellularly, causing vessel destruction, diapedesis, erythrocyte extravasation, fibrin deposition, and, ultimately, tissue necrosis in the region dependent on these vessels for perfusion. It has been postulated that vasculitic lesions are more likely to occur in leg vessels. In an erect position, the elevated hydrostatic pressure in the "dependent parts of the body," as well as tortuosity of these vessels, causes more distorted and turbulent flow patterns and stasis. This predisposes to immune complex and fibrin deposition and resultant destructive inflammatory vasculitis. The resultant vascular inflammation is completely nonspecific in character, with edema, fibrin deposition, and leukocytic infiltration in the affected vessel wall. Exudate may layer the endothelial surface and predispose to intravascular thrombosis or even rupture of the artery. The inflammatory involvement usually extends into the perivascular tissues. If the inflammatory reaction is prolonged, fibroblastic scarring may eventually cause narrowing and sometimes total obliteration of the vascular lumen. The major alterations are in the intima, which is markedly thickened by a cushion of proliferating fibroblasts, myocytes, and foamy macrophages, often leading to luminal narrowing or obliteration. The thickened intima may be infiltrated by scattered neutrophils and mononuclear cells, and the walls of most of these arteries show deposits of immunoglobulin and complement.6 According to the classification scheme of Copeman and Ryan,7 vasculitis can be subcategorized into neutrophilic (leukocytoclastic) vasculitis, lymphocytic vasculitis, and

granulomatous vasculitis. It is frequently impossible to distinguish between these vasculidities clinically; however, the observation of the type of inflammatory infiltrate under light microscopy is usually diagnostic. Sanchez et a].* list many of the clinical syndromes in which leukocytoclastic vasculitis can be seen. The histologic features required to make the diagnosis of acute vasculitis include vessel destruction characterized by transmural inflammatory infiltrate consisting of neutrophils and nuclear fragments, endothelial swelling, fibrin deposition, and often erythrocyte extravasation as seen in our patient (Figs. 1, 2). Zax et al.9 showed an evolution of inflammatory infiltrate from neutrophils to lymphocytes in sequential biopsy specimens obtained from a patient with vasculitis. They argue that it is the natural course of leukocytoclastic vasculitis to be initially a predominantly neutrophilic process and to evolve into a lymphocytic process. Love and Santoro" have suggested that anticardiolipin and lupus anti-coagulant may be important markers for abnormalities that would lead to thrombosis, thrombocytopenia, or abnormal bleeding. In addition, as suggested by Parlevliet et al.," there is evidence to believe that antibodies to components of neutrophil cytoplasm will prove to be of great value in early diagnosis. Specks et aL'* further explain that these anticytoplasmic autoantibodies determined by the immunofluorescence technique are highly specific for Wegener's granulomatosis and other vasculidities. They argue that the sensitivity is dependent on the extent and activity of disease, and serial determinations are valuable in monitoring disease activity. A similar study was done by Adoue et al.I3 Preliminary but extremely valuable data collected by Grau et al. l4 in Switzerland suggest that a particular pattern

10

Sanger et at.

of cytokine changes is associated with vasculitis and that cytokines might be involved in the pathogenesis of vasculidities such as polyarteritis nodosa, Churg and Straus angiitis, and Wegener’s granulomatosis. These cytokines included tumor necrosis factor alpha, interleukin-1 beta, interleukin-2, interferon, and interferon gamma. Further studies could prove these factors to be valuable markers in monitoring disease activity and therapy. Preoperatively and at the time of hemorrhage from the anastomosis, coagulation studies were normal in our patient. The cause of failure and bleeding from the pedicle was secondary to transmural necrosis of the thoracodorsal artery at the anastomosis and disruption of the suture line. Subsequent skin biopsies, taken when the patient was critically ill, demonstrated widespread acute vasculitis with compliment and immunoglobulin deposits. At autopsy, vasculitis involving all internal organs was also present. This would suggest that all vessels are at risk and that free flap transfer in these patients should be undertaken with extreme care, as the potential for involvement of the anastomosis could result in flap loss, as occurred in this case, or result in segmental infarction of a portion of the flap. Perhaps in the near future we can utilize markers such as anticardiolipin, lupus anticoagulant, cytokines, and neutrophi1 anticytoplasmic antibodies for early detection. In addition, the positive results of these tests could be treated as red flags when we consider free flap transfer in patients with suspected disease of autoimmune origin and vasculitis.

REFERENCES I . O’Brien BM, Monison W A Reconstructive Microsurgery. Edinburgh, London, Melbourne, and New York, Churchill Livingstone, 1987, pp 225-234.

2. Robins SL, Cotran RS, Kumar V (eds): Pathologic Basis of Disease, 3rd ed. Philadelphia, W.B. Sanders Company, 1984, pp 523-524. 3. Rich AR, Gregory JE: The experimental demonstration that periarteritis nodosa is a manifestation of hypersensitivity. Bull J Hopkins Hosp 72:65-88, 1943. 4. Dixon FJ, Vazquez JJ, Wiegle WO, Cochrane CG: Pathogenesis of serum sickness. Arch Parhol Lab Med 65:18-28, 1958. 5. Smoller BR, McNutt NS, Contreras F The natural history of vasculitis. What the histology tells us about pathogenesis. Arch Dermatol 126:84-89, 1990. 6. Ryan TJ; The epidermis and its blood supply in varicose disorders of the leg. Trans St. John’s Dermatol Soc (London). 5551-59, 1969. 7. Copeman PWM, Ryan TJ. The problems of classification of cutaneous angiitis with reference to histopathology and pathogenesis. Br J Dermatol 82(suppl):2- 14, 1970. 8. Sanchez NP, Van Hale HM, Su D: Clinical and hisotpathologic spectrum of neurotizing vasculitis. Arch Dermatol 121:220-223, 1985. 9. Zax RH, Hodge SJ. Callen JP. Cutaneous leukocytoclastic vasculitis: Serial histopathologic evaluation demonstrates the dynamic nature of the infiltrate. Arch Dermatol 126:69-72, 1990. 10. Love PE, Santoro SA: Antiphospholipid antibodies: Anticardiolipin and the lupus anticoagulant in systemic lupus erythematosis (SLE) and in non-SLE disorders, prevalence and clinical significance. Ann Intern Med 112:682-698, 1990. 1 1 . Parlevliet KJ, Henzler-Logmans SC, Oe PL, Bronsveld W, Balm AJ, Donker AJ. Antibodies to components of neutrophil cytoplasm: A new diagnostic tool in patients with Wegener’s granulomatosis and systemic vasculitis. Q J Med 6655-63, 1988. 12. Specks U, Wheatley CL, Donald TJ, Rohrbach MS, DeRemee RA. Anticytoplasmic autoantibodies in the diagnosis and follow-up of Wegener’s granulomatosis. Mayo Clin Proc 64:28-36, 1989. 13. Adoue D, Vernier 1, Oksman F. Antibodies directed against the cytoplasm of granulocytes. Diagnostic value in vasculitis. Ann Med Interne 140435-439, 1989. 14. Graue GE, Roux-Lombard P, Gysler C, Lambert C, Lambert PH, Dayer JM, Guillerin L. Serum cytoxine changes in systemic vasculitis. Immunology 68:196-198, 1989.